To evaluate the performance of the American College of Radiology ovarian-adnexal reporting and data system (O-RADS) in the diagnosis of adnexal masses (AMs) by pelvic ultrasound.

This retrospective study included 441 patients with AMs who underwent surgery and had pathological diagnosis at the Maternal and Child Health Hospital of Hubei Province from January 2019 to December 2020. All patients were detected by ultrasound before surgery, and AMs were categorized into four categories according to the O-RADS classification. The value of O-RADS classification for the diagnosis of benign and malignant AMs was then assessed by receiver operating characteristic (ROC) curve analysis. Using histopathology as the reference standard, the diagnostic performance of O-RADS for detecting malignant AMs was calculated. The four-grid table was used to analyze the validity of the system.

A total of 441 AMs were evaluated: 88 were malignant and 353 were benign. Malignant tumors accounted for 1.1%, 3.7%, 47.8%, and 91.1% of O-RADS 2, 3, 4, and 5 AMs, respectively. The area under the ROC of O-RADS classification was 0.947 (95% confidence interval: 0.919-0.975). The best cutoff value for predicting malignant AMs was 3.5. When considering both O-RADS 4 and 5 as predictors of malignancy, the sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and Youden index of O-RADS were 94.3%, 88.6%, 67.5%, 98.4%, 8.27, 0.06, and 0.82, respectively. When considering only O-RADS 5 as a predictor of malignancy, the sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and Yoden index of O-RADS were 57.9%, 98.6%, 91.1%, 90.3%, 41.36, 0.43, and 0.57, respectively.

O-RADS can be used as a reliable method for the differential diagnosis of benign and malignant AMs. Both O-RADS 4 and 5 should be used as predictors of malignancy.